Qwen3.8 27B text-only GGUF model compressed by TurinTech with Artemis. Recommended: Qwen3.8-27B-Artemis-IQ2_XXS.gguf. Model files, checksums and measured trade-offs: https://huggingface.co/turintech/Qwen3.8-27B-Artemis-GGUF. Download size is not runtime memory. Measured IQ2_XXS: Mean KLD 1.391, Same top p 68.6% vs BF16 on All 250 texts, each text weighted equally; protocol rtx5090-2026-10-25.
Qwen3.8 27B text-only GGUF model compressed by TurinTech with Artemis. Recommended: Qwen3.8-27B-Artemis-IQ2_XXS.gguf. Model files, checksums and measured trade-offs: https://huggingface.co/turintech/Qwen3.8-27B-Artemis-GGUF. Download size is not runtime memory. Measured IQ2_XXS: Mean KLD 1.391, Same top p 68.6% vs BF16 on All 250 texts, each text weighted equally; protocol rtx5090-2026-10-25.
Qwen3.8 27B text-only GGUF model compressed by TurinTech with Artemis. Recommended: Qwen3.8-27B-Artemis-IQ2_XXS.gguf. Model files, checksums and measured trade-offs: https://huggingface.co/turintech/Qwen3.8-27B-Artemis-GGUF. Download size is not runtime memory. Measured IQ2_XXS: Mean KLD 1.391, Same top p 68.6% vs BF16 on All 250 texts, each text weighted equally; protocol rtx5090-2026-10-25.
Qwen3.8 27B text-only GGUF model compressed by TurinTech with Artemis. Recommended: Qwen3.8-27B-Artemis-IQ2_XXS.gguf. Model files, checksums and measured trade-offs: https://huggingface.co/turintech/Qwen3.8-27B-Artemis-GGUF. Download size is not runtime memory. Measured IQ2_XXS: Mean KLD 1.391, Same top p 68.6% vs BF16 on All 250 texts, each text weighted equally; protocol rtx5090-2026-10-25.